This disclosure is generally directed to media systems configured to selectively record portions of a media asset. In particular, methods and systems are provided for improved selection of relevant portions of a media asset based on analysis of the audio component data of the media asset.